BCH709 Introduction to Bioinformatics: 6_Introduction of R & R plotting II

Introduction of R & R plotting II

Ph.D. Tong Zhou

Tong Zhou, Ph.D.
L-207B, Center for Molecular Medicine, MS575
Department of Physiology and Cell Biology
University of Nevada, Reno School of Medicine
1664 North Virginia Street, Reno , NV 89557

The Zhou Lab carries out translational and theoretical research in bioinformatics and computational biology. Much of our research addresses questions of computational molecular medicine and molecular evolution, in particular about the use of genomic data to understand the pathobiology and develop biomarkers for human diseases and to understand the mechanism of exnoic sequence evolution.

Introduction to R

R (www.r-project.org) is a commonly used free Statistics software. R allows you to carry out statistical analyses in an interactive mode, as well as allowing simple programming. R

Prepare your laptop

Open two terminal

Connect one to pronghorn

ssh <yourID>@pronghorn.rc.unr.edu

R Installation

Prepare working folder in Pronghorn

cd ~/
mkdir r_plot
cd r_plot
wget -O dataset1.txt https://pastebin.com/raw/N5g8bXg6
wget -O dataset2.txt https://pastebin.com/raw/nAMG57Qy
wget -O heart.txt https://pastebin.com/raw/pN4Tjkkp
wget -O liver.txt https://pastebin.com/raw/Df8vh0Gz
conda activate r_plot

If you missed previous class

cd ~/
mkdir r_plot
cd r_plot
wget -O r_plot.yaml https://pastebin.com/raw/kSAC1AsK
conda env create -n r_plot -f r_plot.yaml
conda activate r_plot

Download results to your laptop

Try this on your laptop

###Please replace <YOURID> to your id.
scp <YOURID>@pronghorn.rc.unr.edu:~/r_plot/*.pdf .

Start R

R

Introduction of R & R plotting – part 2

Matrix

Matrix manipulation

m = matrix(1:9, ncol=3)
m[1,3]			#Access the element in row 1/column 3
m[2,]			#Access row 2
m[,3]			#Access column 3
m[,-2]			#Remove column 2
m[-1,]			#Remove row 1
m[2,3] = 100			#Change the value in row 2/column 3
m[,1] = 200			#Change the value in column 1
m[,2:3] = 201:206		#Change the value in column 2 and 3
m = matrix(1:50, ncol=5)
#We asked R to give us all the values of matrix  _m_  which had values greater than 5, and it returned these values as a mathematical vector
m[m>25]
m[m>25] = -1		 #change all the values in m that are greater than 5 to -1

Several useful functions

Exercise

Generate a matrix with elements from 1 to 80 (10 rows and 8 columns)

m = matrix(1:80, nrow=10)

Change the values in the matrix that are divisible by 3 or 7 to 100

m[m%%3==0 | m%%7==0] = 100

Pick up the rows with row mean > 70

m[rowMeans(m)>70,]

Data frame

Data frame manipulation

#Show column names
colnames(data)
#Use dolloar sign $ to access a column
data$sample
data$gene2
#Pick up the samples with gene1 expression > 5 and gene2 expression <15
data[data$gene1 > 5 & data$gene2 <15,]
#Add a new column to the data frame
gene3 = rnorm(20, mean=2, sd=0.3)
data = data.frame(data, gene3)
#Plot a boxplot showing the difference in gene1 expression between the WT and KO samples
pdf(file="boxplot_gene1.pdf", width=3, height=4)
boxplot(data$gene1 ~ data$sample, col=c("grey40", "grey80"), xlab="", ylab="Expression")
dev.off()

Read a data frame from file

#Load the human subject data sheet: dataset2.txt
#52 subject: 16 healthy controls and 36 patientth glioma
data = read.delim("dataset2.txt")
data
#Use the function table() to count at eacmbination of factor levels
table(data$Type)
table(data$Sex)
table(data$Type, data$Sex)
#Split the data according to subject type anmpute the mean age for each
aggregate(data$Age~data$Type, FUN=mean)
#Plot a boxplot showing the difference in agtween the control and glioma groups
pdf(file="boxplot_age1.pdf", width=3, height=4)
boxplot(data$Age ~ data$Type, col=c("grey80","orange"), xlab="", ylab="Age")
dev.off()

Exercise

Extract the female subjects with age>50

data[data$Sex == "F" & data&Age>50,]

Plot a boxplot showing the group difference in age among the male subjects

data[data$Sex=="F" & data$Age>50,]
pdf(file="boxplot_age2.pdf", width=3, height=4)
male = data[data$Sex=="M",]
boxplot(male$Age ~ male$Type)
dev.off()
#Alternative solution
pdf(file="boxplot_age2.pdf", width=3, height=4)
boxplot(data$Age[data$Sex=="M"] ~ data$Type[data$Sex=="M"])
dev.off()

R plotting

Split display and plot figures

 pdf(file="split_screen.pdf", width=8, height=7)
 split.screen(c(2, 3))	# split display into six sreens
 screen(1)		# prepare screen 1 for output
 plot(1:10, col="red")
 screen(2)		# prepare screen 2 for output
 plot(1:10, col="orange")
 screen(3)		# prepare screen 3 for output
 plot(1:10, col="yellow")
 screen(4)		# prepare screen 4 for output
 plot(1:10, col="green")
 screen(5)		# prepare screen 5 for output
 plot(1:10, col="blue")
 screen(6)		# prepare screen 6 for output
 plot(1:10, col="purple")
 close.screen(all = TRUE)	# exit split-screen mode
 dev.off()

Dataset “Dataset1.txt”

Exercise

Based on “dataset1.txt”, generate boxplots showing the difference in expression of IGSF10 , SFTPH , and FMO2 between normal and tumor tissues

# boxplot showing the difference in expression of IGSF10, SFTPH, and FMO2 between normal and tumor tissues
pdf(file="boxplot_three_genes.pdf", width=8, height=4)
split.screen(c(1, 4))
screen(1)
boxplot(t(expr["IGSF10",])~cl, xlab="", ylab="Expression", main="IGSF10", col=c("grey80", "orange"), xaxt="n")
screen(2)
boxplot(t(expr["SFTPH",])~cl, xlab="", ylab="Expression", main="SFTPH", col=c("grey80", "orange"), xaxt="n")
screen(3)
boxplot(t(expr["FMO2",])~cl, xlab="", ylab="Expression", main="FMO2", col=c("grey80", "orange"), xaxt="n")
screen(4)
legend("left", fill=c("grey80", "orange"), c("Normal", "Tumor"), bty="n")
close.screen(all = TRUE)
dev.off()

par()

Dataset “heart.txt”

Load the “heart.txt” dataset

expr = read.delim("heart.txt", row.names="gene")

Exercise

Solution

#Correlation in expression between linear and circular transcripts in liver
expr = read.delim("liver.txt", row.names="gene")
pdf(file="expression_distribution3.pdf", width=4, height=4)
par(mai=c(0.8, 0.8, 0.3, 0.3), mgp=c(2, 0.5, 0), tck=-0.03, cex.main=1, font.main=1)
plot(density(log10(expr$linear)), col="red", main="Liver", xlab="log10 (TPM)", ylab="Density")
lines(density(log10(expr$circular)), col="blue")
dev.off()

cor.test(expr$linear, expr$circular)
cor.test(expr$linear, expr$circular, method="spearman")
pdf(file="correlation_liver.pdf", width=4, height=4, useDingbats=FALSE)
par(mai=c(0.8, 0.8, 0.3, 0.3), mgp=c(2, 0.5, 0), tck=-0.03, cex.main=1, font.main=1)
 plot(log10(expr$linear), log10(expr$circular), xlim=c(-4, 4), ylim=c(-4, 4), col="grey70", pch=20, cex=0.5, xlab=expression(paste("Mean ", TPM[linear])), ylab=expression(paste("Mean ", TPM[circ])), main="Liver", axes=F)
lines(lowess(log10(expr$linear), log10(expr$circular), f=1/3), col="red", lty=1)
abline(a=0, b=1, lty=2)
axis(1, at=-2:2*2, c(expression(10^-4), expression(10^-2), 1, expression(10^2), expression(10^4)), cex.axis=1)
axis(2, at=-2:2*2, c(expression(10^-4), expression(10^-2), 1, expression(10^2), expression(10^4)), cex.axis=1)
dev.off()

Contact information

Email: tongz@med.unr.edu

If you have any question, please contact me by email.

Reading materials

The R Project for Statistical Computing https://www.r-project.org/
R Graphic cookbook https://learning.oreilly.com/library/view/r-graphics-cookbook/9781491978597/
R workshop https://bioinformatics.ca/workshops/2018-introduction-to-R/
RStudio https://www.rstudio.com/
Wikipedia R (programming language) https://en.wikipedia.org/wiki/R_(programming_language)
Wikipedia RStudio https://en.wikipedia.org/wiki/RStudio